The long-term, programmatic goal of this research is to advance prevention of addictive behaviors especially substance use in normal adolescent psychological development. The overall goal of this proposal is to pilot test a novel, preventive intervention for adolescent substance use. More specifically, it will evaluate the efficacy of yoga taught during school to positively influence risk and protective factors of substance use and the initiation and severity of substance use. Prevalence of substance use among adolescents remains alarmingly high. Existing school-based interventions are primarily educational and have shown limited effectiveness. Recently, skill-building interventions especially targeting individual life skills have been recommended. Yoga likely promotes development of self-regulatory skills while reducing negative internalizing behaviors, which are widely documented precursors of adolescent substance use. Existing research supporting yoga as a substance use intervention is scarce, and is focused on treating abuse in adults. Minimal data exists showing yoga improves psychosocial health and well-being of youth. The proposed research will be the first time yoga is evaluated for intrapersonal predictors of substance use and substance use itself in an adolescent population. Uniquely, the intervention will be delivered during school within the physical education (P.E.) curriculum, allowing the use of P.E. as a control condition. A group randomized, controlled trial will be conducted of yoga classes throughout an academic year during P.E. classes in a public high school. Classes of ninth-graders will be randomly assigned to either yoga during P.E. or P.E.-as-usual. Validated, self-report surveys addressing: negative internalizing behaviors related to mood, stress and impulsivity (Specific Aim 1); emotional and behavioral dimensions of self-regulation (Specific Aim 2); and substance use initiation and severity (Specific Aim 3) will be administered before and after the yoga program to students in both groups. ANCOVA controlling for baseline values will compare scores between the two groups for each outcome. Structural equation modeling will identify mediating variables and putative self-regulatory mechanisms of change by which yoga may prevent substance use in adolescence. These results will be used to develop a mechanistic hypothesis, and apply for additional funding to further evaluate the efficacy of yoga during P.E. for preventing substance use initiation and improving substance use trajectories throughout adolescence.